TL;DR
This paper introduces a novel methodology for evaluating and classifying overseas talents in China by extending the Best-Worst Method (BWM) to handle "intuitionistic preference relations." This approach accounts for both the preferred and non-preferred aspects of decision-maker judgments, offering a more nuanced and practical way to assess talent compared to traditional methods.
What Problem Does This Paper Address?
As China actively seeks to attract and leverage overseas talent for economic development, a robust and systematic method for evaluating and classifying these individuals is crucial. Existing decision-making tools often struggle to capture the inherent uncertainty and dual nature of human preferences, particularly when dealing with complex evaluations like talent assessment. The authors identify a need for a decision-aid that can handle these complexities, considering not just how much one option is preferred over another, but also how much it is not preferred, reflecting real-world decision-making more accurately.
Methods
The core of this paper is the development of a "non-linear Best-Worst Method (BWM) for intuitionistic relations." This method builds upon the original BWM by incorporating "intuitionistic preference relations," which represent judgments as pairs of preferred and non-preferred degrees.
Here's a breakdown of the key methodological contributions:
- Intuitionistic Preference Relations: Instead of simple numerical comparisons, decision-makers express their preferences using intuitionistic multiplicative numbers (IMNs), which include both a preferred degree (e.g., how much better A is than B) and a non-preferred degree (e.g., how much worse A is than B). This captures the inherent uncertainty and hesitation in human judgment.
- Geometry Consistency: A novel approach to determine the consistency of decision-maker judgments is introduced, based on geometric consistency. This helps in calculating the weights of both decision-makers and criteria.
- Non-linear BWM: The standard BWM is adapted to work with these intuitionistic preference relations. A key innovation is the consideration of both positive and negative aspects of preferences simultaneously, leading to a more comprehensive evaluation. This involves minimizing deviations from the ideal "best" and "worst" comparisons, accounting for both preferred and non-preferred degrees.
- Fuzzy-Based Geometric Index: A transformation formula is used to convert intuitionistic multiplicative numbers into intuitionistic fuzzy numbers, allowing for integration with fuzzy set theory and facilitating calculations.
Datasets
The paper uses an illustrative example based on real-world data from the selection process for the Tian Jin Haihe Friendship Award in 2015. This dataset involves 20 overseas talents evaluated by a group of three experts across five criteria: basic quality, capacity, contribution, development potential, and internationalization. The authors state that this data is part of an ongoing project led by a relevant department in Tianjin.
Key Insights and Ideas
- The Importance of "Non-Preferred" Degrees: The study highlights that simply knowing how much someone prefers option A over option B isn't enough. Understanding how much they dislike option A compared to B provides crucial additional information for accurate decision-making, especially in complex scenarios like talent evaluation.
- Intuitionistic Relations Mirror Human Judgment: The use of intuitionistic preference relations better reflects the nuanced and often uncertain nature of human decision-making, making the evaluation process more realistic and practical.
- BWM's Adaptability: The paper demonstrates that the BWM framework can be effectively extended to handle more complex preference structures, such as intuitionistic fuzzy sets, without losing its core efficiency benefits.
- Categorization for Targeted Management: The classification of talents into different groups based on their evaluation scores suggests a practical application for differentiated management strategies, such as tailored services for different talent tiers.
Key Results
The proposed methodology was applied to an illustrative example of evaluating 20 overseas talents. The results showed that the method could effectively rank alternatives and classify them into different categories. The study demonstrated that considering both positive and negative preference aspects led to different rankings, underscoring the value of the intuitionistic approach. The authors also presented consistency measures (VD and DD) to assess the reliability of the decision-makers' inputs.
Limitations and Future Work
The authors acknowledge several limitations:
- Hesitation Degree: While intuitionistic relations capture some hesitation, they may still be insufficient for describing decision-makers' hesitation in highly complex situations.
- Incomplete Information: The current model does not explicitly focus on decision-making with incomplete information, which is a common challenge in real-world scenarios.
- Moderate Scale: The method is suitable for a moderate number of criteria and alternatives. For very large-scale problems, the workload might become substantial.
- Consistency Degree: The paper focuses on geometry consistency for weighting but notes that the consistency of the BWM itself with intuitionistic preference relations has not been fully explored and could influence ranking results.
Future research directions suggested include:
- Extending the methodology to interval intuitionistic preference relations.
- Exploring methods to handle missing elements in decision matrices.
- Investigating the consistency degree of the BWM for intuitionistic preference relations.
- Potentially assigning secondary weights to decision-makers to refine results.
- Applying the methodology to other fields beyond talent evaluation, such as talent recruitment, investment decision-making, and supplier selection.
Summary updated